1
00:00:00,930 --> 00:00:02,720
- [Instructor] It's relatively common

2
00:00:02,720 --> 00:00:05,580
in Python to have
functions that can receive

3
00:00:05,580 --> 00:00:08,730
an arbitrary number of arguments.

4
00:00:08,730 --> 00:00:11,170
In fact, we've already used a couple

5
00:00:11,170 --> 00:00:13,420
of those functions in Python.

6
00:00:13,420 --> 00:00:16,020
For example, we used the min function

7
00:00:16,020 --> 00:00:19,220
to be able to do things like this

8
00:00:19,220 --> 00:00:22,360
where we could pass any
number of arguments we want

9
00:00:22,360 --> 00:00:25,930
and it will figure out the
minimum of those arguments.

10
00:00:25,930 --> 00:00:30,930
Now, the way that you define
such a function in Python

11
00:00:30,950 --> 00:00:35,500
is by using an asterisk
on a parameter name.

12
00:00:35,500 --> 00:00:37,850
So let's show you what
we're talking about here.

13
00:00:37,850 --> 00:00:41,120
So let's say we want to
define an average function

14
00:00:41,120 --> 00:00:45,640
that's capable of taking
the average of any number

15
00:00:45,640 --> 00:00:49,490
of numeric values that you
pass into the function.

16
00:00:49,490 --> 00:00:52,210
You can write a parameter of this form,

17
00:00:52,210 --> 00:00:56,410
where the word args is just a
convention, not a requirement,

18
00:00:56,410 --> 00:01:00,480
and let me go ahead and write
out the rest of the function

19
00:01:00,480 --> 00:01:02,190
and then we'll talk about this.

20
00:01:02,190 --> 00:01:05,410
So we want to return
the sum of the arguments

21
00:01:05,410 --> 00:01:10,410
and we want to divide that by
the length of that parameter,

22
00:01:10,450 --> 00:01:13,500
which is going to be the
number of total arguments.

23
00:01:13,500 --> 00:01:16,660
So what happens when
you apply the asterisk

24
00:01:16,660 --> 00:01:21,370
to a parameter name is this
tells the Python interpreter

25
00:01:21,370 --> 00:01:23,600
when somebody calls this function,

26
00:01:23,600 --> 00:01:28,290
pack up all of the arguments into a tuple

27
00:01:28,290 --> 00:01:32,640
and pass that tuple into
the function definition.

28
00:01:32,640 --> 00:01:35,240
So args is actually going to be a tuple

29
00:01:35,240 --> 00:01:38,820
of whatever arguments are
passed to this function.

30
00:01:38,820 --> 00:01:41,770
Now, we've seen previously
that when you pass a sequence

31
00:01:41,770 --> 00:01:44,540
to the sum function, it
can calculate the sum

32
00:01:44,540 --> 00:01:46,650
of all the numbers in that sequence

33
00:01:46,650 --> 00:01:48,710
and we've seen that
when you pass a sequence

34
00:01:48,710 --> 00:01:50,640
to the len function, it can tell you

35
00:01:50,640 --> 00:01:53,540
how many elements are in that sequence.

36
00:01:53,540 --> 00:01:56,280
So let's go ahead and define that function

37
00:01:56,280 --> 00:01:59,540
and now we can go and
calculate the average

38
00:01:59,540 --> 00:02:00,840
of a couple of values.

39
00:02:00,840 --> 00:02:03,180
So let's say we give it five and 10,

40
00:02:03,180 --> 00:02:07,370
and it gives me back the average of 7.5.

41
00:02:07,370 --> 00:02:09,980
Let's recall that and
add in another value,

42
00:02:09,980 --> 00:02:12,410
and of course the average
of these three values

43
00:02:12,410 --> 00:02:13,830
happens to be 10

44
00:02:13,830 --> 00:02:16,400
and let's recall it and
add in another value

45
00:02:16,400 --> 00:02:18,640
just to show you that, once again,

46
00:02:18,640 --> 00:02:21,640
we can calculate the
average of those values.

47
00:02:21,640 --> 00:02:24,850
Now, I'm not doing any
error checking up here

48
00:02:24,850 --> 00:02:28,160
so it is possible I can get
an error with this as well.

49
00:02:28,160 --> 00:02:31,210
For example, if I call
average with no arguments,

50
00:02:31,210 --> 00:02:35,000
then it's going to result
in a zero division error

51
00:02:35,000 --> 00:02:36,940
because of the fact that I didn't check

52
00:02:36,940 --> 00:02:38,370
for divide by zero.

53
00:02:38,370 --> 00:02:41,340
If the length of the tuple is zero,

54
00:02:41,340 --> 00:02:43,490
there are no elements and therefore,

55
00:02:43,490 --> 00:02:46,880
I wind up with the error
that you see down below here.

56
00:02:46,880 --> 00:02:50,170
Now, if you want to use this capability,

57
00:02:50,170 --> 00:02:52,330
you can do that in any function,

58
00:02:52,330 --> 00:02:54,500
including functions with arguments,

59
00:02:54,500 --> 00:02:56,620
with required arguments or arguments

60
00:02:56,620 --> 00:02:58,040
that have default values,

61
00:02:58,040 --> 00:03:01,290
but no matter what,
this particular argument

62
00:03:01,290 --> 00:03:04,963
must be the last one
in the parameter list.

63
00:03:06,510 --> 00:03:09,590
Now, one interesting thing about Python

64
00:03:09,590 --> 00:03:14,590
is an operator that enables you
to take a sequence of values

65
00:03:14,890 --> 00:03:19,090
and unpack that sequence
so that the values

66
00:03:19,090 --> 00:03:21,580
can be passed into a function.

67
00:03:21,580 --> 00:03:23,960
So, for example, the average function

68
00:03:23,960 --> 00:03:26,030
that we defined up above here

69
00:03:26,030 --> 00:03:29,560
is capable of receiving a
whole bunch of arguments

70
00:03:29,560 --> 00:03:32,610
that perhaps are stored in a list already.

71
00:03:32,610 --> 00:03:36,247
So, as we demonstrated
in some earlier examples,

72
00:03:36,247 --> 00:03:39,330
if you want to calculate a class average,

73
00:03:39,330 --> 00:03:42,650
you are going to total up some grades

74
00:03:42,650 --> 00:03:45,590
and then divide by the number of students.

75
00:03:45,590 --> 00:03:49,510
Well, let's assume that we
already have a list of grades.

76
00:03:49,510 --> 00:03:51,200
Let's call it grades,

77
00:03:51,200 --> 00:03:53,500
and let's go ahead and
put some values in there,

78
00:03:53,500 --> 00:03:56,750
so let's do 88 and 75

79
00:03:56,750 --> 00:04:00,730
and 96 and 55 and 83.

80
00:04:00,730 --> 00:04:03,100
So, of course, we've already seen

81
00:04:03,100 --> 00:04:05,930
that we can do things
like call the sum function

82
00:04:05,930 --> 00:04:08,560
with grades as an argument and calculate

83
00:04:08,560 --> 00:04:10,450
the sum of these values.

84
00:04:10,450 --> 00:04:13,330
We've also seen that we
can use the len function

85
00:04:13,330 --> 00:04:15,980
to figure out the number
of values in this list

86
00:04:15,980 --> 00:04:19,310
and, of course, we can use
that to calculate the average

87
00:04:19,310 --> 00:04:22,170
but we already have that defined up here

88
00:04:22,170 --> 00:04:24,230
in our average function.

89
00:04:24,230 --> 00:04:26,400
So, what we can do instead

90
00:04:26,400 --> 00:04:30,287
is call the average
function and we can say,

91
00:04:30,287 --> 00:04:34,367
"Unpack the elements of
the list called grades

92
00:04:34,367 --> 00:04:37,710
"and pass those into
the average function."

93
00:04:37,710 --> 00:04:41,960
So this is the equivalent
of literally passing in

94
00:04:41,960 --> 00:04:45,180
this highlighted,
comma-separated list of values

95
00:04:45,180 --> 00:04:48,500
as the set of arguments
to the average function

96
00:04:48,500 --> 00:04:52,773
and indeed, it gives me back
the class average in that case.